C++ Document::set_data方法代码示例

void XapianIndex::setDocumentData(Xapian::Document &doc, const DocumentInfo &info,
	const string &language) const
{
	string title(info.getTitle());
	string timestamp(info.getTimestamp());
	char timeStr[64];

	// Set the document data omindex-style
	string record = "url=";
	record += info.getLocation();
	// The sample will be generated at query time
	record += "\nsample=";
	record += "\ncaption=";
	if (badField(title) == true)
	{
		// Modify the title if necessary
		string::size_type pos = title.find("=");
		while (pos != string::npos)
		{
			title[pos] = ' ';
			pos = title.find("=", pos + 1);
		}
#ifdef DEBUG
		cout << "XapianIndex::setDocumentData: modified title" << endl;
#endif
	}
	record += title;
	record += "\ntype=";
	record += info.getType();
	// Append a timestamp
	record += "\ntimestamp=";
	record += timestamp;
	// ...and the language
	record += "\nlanguage=";
	record += language;
#ifdef DEBUG
	cout << "XapianIndex::setDocumentData: document data is " << record << endl;
#endif
	doc.set_data(record);

	// Add this value to allow sorting by date
	snprintf(timeStr, 64, "%d", TimeConverter::fromTimestamp(timestamp));
	doc.add_value(0, timeStr);
}

Indexer::Indexer(const string &datapath, const string &dbpath)
{
    // Hardcode field offsets for simplicity.
    const size_t FIELD_ID_NUMBER = 0;
    const size_t FIELD_TITLE = 2;
    const size_t FIELD_DESCRIPTION = 8;

    // Create or open the database we're going to be writing to.
    Xapian::WritableDatabase db(dbpath, Xapian::DB_CREATE_OR_OPEN);

    // Set up a TermGenerator that we'll use in indexing.
    Xapian::TermGenerator termgenerator;
    termgenerator.set_stemmer(Xapian::Stem("en"));

    ifstream csv(datapath.c_str());
    vector<string> fields;
    csv_parse_line(csv, fields);

    // Check the CSV header line matches our hard-code offsets.
    if (fields.at(FIELD_ID_NUMBER) != "id_NUMBER" ||
    fields.at(FIELD_TITLE) != "TITLE" ||
    fields.at(FIELD_DESCRIPTION) != "DESCRIPTION") {
    // The CSV format doesn't match what we expect.
    cerr << "CSV format has changed!" << endl;
    exit(1);
    }

    while (csv_parse_line(csv, fields)) {
    // 'fields' is a vector mapping from field number to value.
    // We look up fields with the 'at' method so we get an exception
    // if that field isn't set.
    //
    // We're just going to use DESCRIPTION, TITLE and id_NUMBER.
    const string & description = fields.at(FIELD_DESCRIPTION);
    const string & title = fields.at(FIELD_TITLE);
    const string & identifier = fields.at(FIELD_ID_NUMBER);

    // We make a document and tell the term generator to use this.
    Xapian::Document doc;
    termgenerator.set_document(doc);

    // Index each field with a suitable prefix.
    termgenerator.index_text(title, 1, "S");
    termgenerator.index_text(description, 1, "XD");

    // Index fields without prefixes for general search.
    termgenerator.index_text(title);
    termgenerator.increase_termpos();
    termgenerator.index_text(description);

    // Store all the fields for display purposes.
    doc.set_data(identifier + "\n" + title + "\n" + description);

    // We use the identifier to ensure each object ends up in the
    // database only once no matter how many times we run the
    // indexer.
    string idterm = "Q" + identifier;
    doc.add_boolean_term(idterm);
    db.replace_document(idterm, doc);
    }
}

int main(int argc, char **argv)
{
    if(argc < 2) {
        usage(argv);
        return 1;
    }

    try {
        char *action = argv[1];
        char *db_path = argv[2];

        if(!strcmp(action, "index")) {
            Xapian::WritableDatabase db(db_path, Xapian::DB_CREATE_OR_OPEN);

            Xapian::TermGenerator indexer;
            Xapian::Stem stemmer("english");
            indexer.set_stemmer(stemmer);

            std::string doc_txt;
            while(true) {
                if(std::cin.eof()) break;

                std::string line;
                getline(std::cin, line);
                doc_txt += line;
            }

            if(!doc_txt.empty()) {
                Xapian::Document doc;
                doc.set_data(doc_txt);

                indexer.set_document(doc);
                indexer.index_text(doc_txt);

                db.add_document(doc);

                std::cout << "Indexed: " << indexer.get_description() << std::endl;
            }

            db.commit();
        } else if(!strcmp(action, "search")) {
            if(argc < 4) {
                std::cerr << "You must supply a query string" << std::endl;
                return 1;
            }

            Xapian::Database db(db_path);
            Xapian::Enquire enquire(db);

            std::string query_str = argv[3];
            argv+= 4;
            while(*argv) {
                query_str += ' ';
                query_str += *argv++;
            }

            Xapian::QueryParser qp;
            Xapian::Stem stemmer("english");
            qp.set_stemmer(stemmer);
            qp.set_database(db);
            qp.set_stemming_strategy(Xapian::QueryParser::STEM_SOME);

            Xapian::Query query = qp.parse_query(query_str);
            std::cout << "Parsed query is: " << query.get_description() <<
                         std::endl;

            enquire.set_query(query);
            Xapian::MSet matches = enquire.get_mset(0, 10);

            std::cout << matches.get_matches_estimated() << " results found.\n";
            std::cout << "Matches 1-" << matches.size() << ":\n" << std::endl;

            for (Xapian::MSetIterator i = matches.begin();
                    i != matches.end(); ++i) {
                std::cout << i.get_rank() + 1 << ": " << i.get_percent() <<
                        "% docid=" << *i << " [" <<
                        i.get_document().get_data()<< "]" << std::endl <<
                        std::endl;
            }
        } else {
            std::cerr << "Invalid action " << action << std::endl;
            usage(argv);
            return 1;
        }

    } catch (const Xapian::Error &error) {
        std::cout << "Exception: " << error.get_msg() << std::endl;
    }
}

bool
DatabaseWrite::rebuild (GList *cpt_list)
{
	string old_path = m_dbPath + "_old";
	string rebuild_path = m_dbPath + "_rb";
	string db_locale;

	// Create the rebuild directory
	if (g_mkdir_with_parents (rebuild_path.c_str (), 0755) != 0) {
		g_warning ("Unable to create database rebuild directory.");
		return false;
	}

	// check if old unrequired version of db still exists on filesystem
	if (g_file_test (old_path.c_str (), G_FILE_TEST_EXISTS)) {
		g_warning ("Existing xapian old db was not cleaned previously: '%s'.", old_path.c_str ());
		as_utils_delete_dir_recursive (old_path.c_str ());
	}

	// check if old unrequired version of db still exists on filesystem
	if (g_file_test (rebuild_path.c_str (), G_FILE_TEST_EXISTS)) {
		g_debug ("Removing old rebuild-dir from previous database rebuild.");
		as_utils_delete_dir_recursive (rebuild_path.c_str ());
	}

	Xapian::WritableDatabase db (rebuild_path, Xapian::DB_CREATE_OR_OVERWRITE);

	Xapian::TermGenerator term_generator;
	term_generator.set_database(db);
	try {
		/* this tests if we have spelling suggestions (there must be
		 * a better way?!?) - this is needed as inmemory does not have
		 * spelling corrections, but it allows setting the flag and will
		 * raise a exception much later
		 */
		db.add_spelling("test");
		db.remove_spelling("test");

		/* this enables the flag for it (we only reach this line if
		 * the db supports spelling suggestions)
		 */
		term_generator.set_flags(Xapian::TermGenerator::FLAG_SPELLING);
	} catch (const Xapian::UnimplementedError &error) {
		// Ignore
	}

	for (GList *list = cpt_list; list != NULL; list = list->next) {
		AsComponent *cpt = (AsComponent*) list->data;

		Xapian::Document doc;
		term_generator.set_document (doc);

		doc.set_data (as_component_get_name (cpt));

		// Sanity check
		if (!as_component_has_install_candidate (cpt)) {
			g_warning ("Skipped component '%s' from inclusion into database: Does not have an installation candidate.",
					   as_component_get_id (cpt));
			continue;
		}

		// Package name
		gchar **pkgs = as_component_get_pkgnames (cpt);
		if (pkgs != NULL) {
			gchar *pkgs_cstr = g_strjoinv (";", pkgs);
			string pkgs_str = pkgs_cstr;
			doc.add_value (XapianValues::PKGNAMES, pkgs_str);
			g_free (pkgs_cstr);

			for (uint i = 0; pkgs[i] != NULL; i++) {
				string pkgname = pkgs[i];
				doc.add_term("AP" + pkgname);
				if (pkgname.find ("-") != string::npos) {
					// we need this to work around xapian oddness
					string tmp = pkgname;
					replace (tmp.begin (), tmp.end (), '-', '_');
					doc.add_term (tmp);
				}
				// add packagename as meta-data too
				term_generator.index_text_without_positions (pkgname, WEIGHT_PKGNAME);
			}
		}

		// Source package name
		const gchar *spkgname_cstr = as_component_get_source_pkgname (cpt);
		if (spkgname_cstr != NULL) {
			string spkgname = spkgname_cstr;
			doc.add_value (XapianValues::SOURCE_PKGNAME, spkgname);
			if (!spkgname.empty()) {
				doc.add_term("AP" + spkgname);
				if (spkgname.find ("-") != string::npos) {
					// we need this to work around xapian oddness
					string tmp = spkgname;
					replace (tmp.begin (), tmp.end (), '-', '_');
					doc.add_term (tmp);
				}
				// add packagename as meta-data too
				term_generator.index_text_without_positions (spkgname, WEIGHT_PKGNAME);
			}
		}
//.........这里部分代码省略.........